The INL is a U.S. Department of Energy National Laboratory operated by Battelle Energy Alliance INL/JOU-16-40271-Revision-0 Desensitizing ignition of energetic materials when exposed to accidental fire Ronald J. Heaps, Michael A. Daniels, Kade Poper, Billy R. Clark, Michelle L. Pantoya August 2015
18
Embed
Desensitizing ignition of energetic materials when exposed ...
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
The INL is a U.S. Department of Energy National Laboratoryoperated by Battelle Energy Alliance
INL/JOU-16-40271-Revision-0
Desensitizing ignition ofenergetic materials whenexposed to accidentalfire
Ronald J. Heaps, Michael A. Daniels,Kade Poper, Billy R. Clark, Michelle L.Pantoya
August 2015
INL/JOU-16-40271-Revision-0
Desensitizing ignition of energetic materials whenexposed to accidental fire
Ronald J. Heaps, Michael A. Daniels, Kade Poper, Billy R. Clark, Michelle L.Pantoya
August 2015
Idaho National LaboratoryIdaho Falls, Idaho 83415
http://www.inl.gov
Prepared for theU.S. Department of EnergyOffice of Nuclear Energy
Under DOE Idaho Operations OfficeContract DE-AC07-05ID14517
1 | P a g e
Desensitizing Ignition of Energetic Materials when Exposed to Accidental Fire
Mr. Kade Poper1, Mr. Billy R. Clark1, Dr. Michelle L. Pantoya1*, Mr. Ronald Heaps, Mr. Michael A. Daniels2
1Texas Tech University, Department of Mechanical Engineering, Lubbock, TX 79409, USA
2Idaho National Laboratory, PO Box 1625, Idaho Falls, ID 83415
Stoichiometry is defined in terms of equivalence ratio (ER) and is the ratio of
fuel/oxidizer mass ratio in the actual mixture to the fuel/oxidizer mass ratio in a stoichiometric
mixture (see Eq. (1)). In this way, mixtures with ER > 1.0 are fuel rich (e.g. for stoichiometric
ER = 1.0).
Once proportioned, the reactants were suspended in hexanes and sonicated in a Misonix
S3000 sonicator for a total of one minute in ten second intervals. Sonication has been shown to
be effective for producing homogeneous composites [10]. Post sonication, the mixtures were
poured into a Pyrex® dish and the hexane evaporated while in a fume hood. The mixed powder
was then reclaimed for further experimentation.
Mixtures were prepared for ER ranging from 1.0 - 5.5. For each ER, two mixtures were
prepared: (1) Al+CuO+CNT (i.e., baseline mixture); and (2) Al+CuO+CNT+AN (i.e., AN
additive mixture) such that the AN additive replacing a portion of CuO could be compared to the
baseline mixture without AN.
2.2 Experimental Methods
Three stages of experimentation included: evaluating combustion pre-heat treatment,
exposing samples to heat treatment simulating accidental fire, and evaluating combustion post-
heat treatment. In all stages experiments were performed in triplicate to establish repeatability.
In the first stage, a 50 mg powder sample for each ER was ignited with a hot wire and the
combustion was recorded using a high speed camera (see Fig. 1 for schematic of setup). A
Nichrome wire is a common hot wire ignition source and provides in excess of 106 degrees per
6 | P a g e
minute heating rate stimulus [24]. A variable voltage source was used to apply 15 volts to the
Nichrome wire in order to achieve the temperature required for ignition. A Phantom v7 (Vision
Research) high speed camera was used to record the combustion event using an F-Stop of 25 and
captured images at 10,000 frames per second.
Figure 1. Experimental setup including high speed camera, blast chamber housing
sample and Nichrome wire ignition system.
The second stage of experimentation exposed each 50 mg sample to simulated accidental
fire conditions using a vacuum oven (NeyTech Qex) in an air environment. The samples were
heated at 10 degrees per minute from room temperature to 230°C and held at this temperature for
1 hour, then cooled to room temperature. An InstruNet (model 100) data acquisition board and
InstruNet software were used to record temperature. This pre-heat temperature is purposefully
above the decomposition temperature of AN so that the effects of AN decomposition on
combustion could be evaluated. While varying the heating rate to simulate a variety of fire
Loose
Powder
Sample
Variable
Voltage
Source
Nichrome
Wire
IgniterHigh Speed
Camera Software
Phantom 7
High Speed
Camera
Blast Chamber
7 | P a g e
exposure conditions was considered, these initial tests were performed for one heating rate to
establish the feasibility of this approach.
In the third stage, post-heat treatment, 50 mg samples were ignited using Nichrome wire
and combustion recorded with the high speed camera at the same operating conditions prior to
heat treatment and shown in Fig. 1. Post-heat treatment, the optimum sample stoichiometries did
not ignite to the point that they could maintain self-sustained energy propagation.
As a further evaluation of these optimum stoichiometries, pre- and post-heat treated
samples were examined for their ability to ignite another energetic mixture using a flame tube
apparatus. In this setup, the tube is 10 cm long with 5 mm inside diameter (see Fig. 2). This
apparatus is commonly used to quantify one-dimensional energy propagation in terms of flame
speed for powder CEM [24, 25]. For this test, half of the tube was filled with 125 mg of pre- or
post-heat treated sample and the other half was filled with 125 mg of an ignition sensitive
mixture of nano-scale particles of aluminum and molybdenum trioxide (Al + MoO3) [26]. This is
one of the most ignition sensitive mixtures and a conservative evaluation of the pre- and post-
heat treated sample ability to initiate a secondary reaction.
Post-heat treated Al+CuO+CNT+AN
Non-ignited Al+MoO3
8 | P a g e
Figure 2. Flame tube apparatus shown AFTER an experiment in which the post-heat treated primer formulation with an ER = 4.0 was ignited but did not ignite the highly ignition sensitive Al+MoO3 powder thermite. An enlarged view of the tube at the junction of the two powders is shown as an inset.
3.0 Results and Discussion
Table 1 summarizes the results. The most interesting finding is that an ER = 4.0 is a
threshold for activation of the decomposition mechanism that was designed to inert the entire
mixture post-heat treatment. This is the ER for which unsustained propagation and a non-ignition
was repeatedly observed.
Table 1. Results of response to Al+CuO+CNT+AN formulations pre- and post-heat treatment as a function of equivalence ratio (ER). Notes provide more perspective on observations.ER Pre-Heat
Treatment Ignition
Post-Heat TreatmentIgnition
Notes
1.6 YES N/A Ignited during bake
1.7 YES YES
1.8 YES YES
2.2 YES YES
2.3 YES YES
2.3(AN Only) YES NO Complete AN decomposition preventing post-heat treatment ignition
3.0 YES YES
3.5 YES NO/YES Non-repeatable results
4.0 YES NO Small amount of propagation but not self-sustained
4.5 YES NO Almost no propagation
5.0 YES NO No propagation but entire 50mg sample was red hot and turned to ash
5.5 YES NO Similar to 5.0 but powder pile exhibited slower heating
9 | P a g e
6.0 YES NO Similar to 5.5 but even slower. No visible flame.
All of the samples that included AN demonstrated comparable visual combustion to the
baseline Al+CuO+CNT at the corresponding equivalence ratio pre-heat treatment. AN
effectively replaced CuO in 1:1 molar ratios and maintained comparable combustion behavior.
Figure 3 shows still frame images of ER = 4.0 threshold case. Notice that the baseline mixture
(see Fig. 3A) and the AN additive mixture (Fig. 3B) demonstrate similar reactivity prior to heat
treatment. However, post-heat treatment (Fig. 3C), the mixture does not achieve a self-sustained
reaction.
10 | P a g e
A
B
C
Figure 3. Representative still frame images of A. Al+CuO+CNT reaction; B. Al+CuO+CNT+AN reaction pre-heat treatment; and, C. Al+CuO+CNT+AN reaction post-heat treatment, all at ER = 4.0
Thermal chemical calculations for the above reactions were performed using REAL code
simulation software (Timtec L.L.C.) for constant specific volume of 0.001 m3/kg and an internal
energy equal to zero. Both adiabatic flame temperature (Fig. 4A) and heat of combustion (Fig.
4B) as a function of equivalence ratio ranging from 1.0-5.5 were simulated. In the post-heat
treated simulations the assumption is that AN does not participate in the reaction such that the
products H2O and N2 do not exist. The simulations indicate that post-heat treatment
decomposition of AN renders the reaction excessively fuel rich such that flame temperatures
drop below the limit for a self-sustaining propagation, identified as 2000K [27, 28]. In fact, for
4.0 ER, the flame temperature drops just below the 2000K limit corresponding well with our
experimental observations of limited reactivity for that formulation (see Fig. 3C). AT ER = 3.5
results are unrepeatable (Table 1), possibly because the 2000 K threshold for energy propagation
is just barely achieved theoretically. Pre-heat treatment flame temperatures and heats of
combustion are comparable for all formulations examined, such that these simulations are also an
indication that AN does not significantly reduce the reactivity of the baseline mixture.
12 | P a g e
A .B.
Figure 4. REAL code simulations for the reaction in Eq. (1) with varying stoichiometry from 1.0 to 5.5 equivalence ratio (i.e., Phi). Solid bars represent pre-heat treatment and hatched bars represent post-heat treatment. Post-heat treatment simulations (i.e., hatched bars) assume no AN in reactants and no H2O or N2 in products. A. adiabatic flame temperature; and, B. heat of combustion.
Optimum stoichiometries (i.e., ER = 4.0 and 4.5) that passed the pre-heat treatment test
and showed non-ignition post-heat treatment, were further tested as primer candidates using the
flame tube apparatus described in Fig. 2. Pre-heat treatment Al+CuO+CNT+AN successfully
ignited the Al+MoO3 mixture. Post-heat treatment, the Al+MoO3 could not ignite. This
assessment was performed for multiple tests to establish repeatability.
4.0 Conclusions
This study presents a new way to tailor energetic material reactants toward their safer
functionality. Aluminum (Al) and copper oxide (CuO) powders combined with 4 vol % carbon
nanotubes (CNT) was the baseline mixture considered. Ammonium nitrate (AN) as an additive to
Al+CuO+CNT was examined by replacing a portion of CuO with equal mole fractions of AN up
13 | P a g e
to 50%. This study explored low temperature AN decomposition as a mechanism to inactivate
the Al+CuO+CNT+AN reaction when exposed to heating conditions simulating accidental fire.
This was a proof of concept study that established a 4.0 equivalence ratio threshold when 50% of
CuO is replaced with AN to cause a non-ignition after heat treatment exposure. The simulated
fire conditions were achieved using a vacuum oven operating in an air environment with
programmed heating at 100 degrees per minute to 230°C. Thermal chemical software was also
used to predict the adiabatic flame temperature and heat of combustion pre- and post- heat
treatment. Results from the simulations are in excellent agreement with the experimental
observations that show unsustained propagation for reactions that produce an adiabatic flame
temperature below 2000 K, an established limit for self-sustained energy propagation for
energetic composites. This mechanism may be extended to higher heating rates as long as the
conditions allow enough time for AN to fully decompose before the composite reaches its
ignition temperature. This study is the first to explore a means to increase the fire safety of a
composite energetic material in any field application.
Acknowledgements
The authors M. Pantoya, B. Clark and K. Poper are grateful for support from the Army Research
Office contract number W911NF-11-1-0439 and encouragement from our program manager, Dr.
Ralph Anthenien. Idaho National Laboratory is also gratefully acknowledged for supporting this
collaborative work with internal funds via the LDRD program. Matt Simmons at Texas Tech
University is gratefully acknowledged for the graphical abstract cover art.
14 | P a g e
References
[1] Fischer SH, Grubelich MC. Theoretical energy release of thermites, intermetallics, and
combustible metals, Sandia National Laboratory Report No.: SAND--98-1176C; CONF-
980728--; Other: ON: DE98005512; BR: 400403310; TRN: US200223%%671, 1998.
[2] Urbanski, T., Chemistry and Technology of Explosives, Pergamon Press, New York, NY,
389-391, 1964.
[3] Teipel, U., Energetic Materials: Processing and Characterization of Particles, Wiley – VCH
Publisher, 2000.
[4] Staley, C. S., Raymond, K. E., Thiruvengadathan, R., Apperson, S. J., Gangopadhyay, K.,
Swaszek, S. M., Taylor, R. J., Gangopadhyay, S., Fast-impulse nanothermite solid-
propellant miniaturized thrusters, Journal of Propulsion and Power 29(6), 1400-1409,
2013.
[5] Francisco, A., Delgado, A., Frias, J., Rubio, M., White, C., Narayana, S., Ashvin, K.,
Shafirovich, E., Combustion of Thermites in Reduced Gravity for Space Applications,
Journal of Thermophysics and Heat Transfer 27(3), 576-583, 2013.